How do Web clients locate Web servers?

Document, resources and services on the web are addressed through a scheme called the Uniform Resource Locator (URL). It is through URLs that clients access the servers. Although similar, URL is not quite the same thing as domain name. URL contains a comprehensive set of information, part of which is the domain name. URL consists of:

The domain name of the computer that holds the information/resource/service,

The port number of the service. Since a number of applications may be in use a port number is used to direct a message to a particular application. This port number could be omitted in which case port number 80 is assumed which is a default of HTTP.

A presentation of a URL could be:

http://www.anywhere.com:80/dir/subdir/file.htm

• The first part of the URL (here HTTP) specifies the service. HTTP is a web service. Other examples of service could be FTP, GOPHER, TELNET, NEWS ...for other services.
• Then follows the name of the computer where service resides or the domain name. In this case www.anywhere.com. Off course this is the name given to the computer simply because IP addresses are not easy to memorise. This domain name is resolved to its corresponding IP address through the use of domain name service.
• Domain name is followed by the port number which is 80 in this case.
• The rest of URL specifies the location for the service on the host computer i.e. the directory and the sub directies (if applicable) and the name of the service.

What are the FTP and GOPHER services?

Food for thought:

How can content of Web server log files be used?

What are the implications of using Web server logs when we have distributed systems?

How do applications convert domain names into IP addresses?